How to Back Up and Restore a Kubernetes Cluster with Velero

知識庫

Velero provides comprehensive backup and restore capability for Kubernetes clusters — covering both resource definitions and persistent volume data. This guide covers setting it up on a self-hosted VPS-based cluster.

Why Kubernetes Needs Dedicated Backup Tooling

Simply backing up etcd (Kubernetes' underlying data store) doesn't capture everything meaningfully — Velero provides application-consistent backup of both Kubernetes resources (deployments, services, configmaps) and associated persistent volume data, plus selective/namespace-scoped restore capability.

Prerequisites

A working Kubernetes cluster (see How to Install kubeadm and Set Up a Multi-Node Kubernetes Cluster)
S3-compatible object storage for backup destination (see How to Set Up Self-Hosted S3-Compatible Object Storage with MinIO for self-hosted option)

Step 1 — Install the Velero CLI

curl -fsSL -o velero.tar.gz https://github.com/vmware-tanzu/velero/releases/download/v1.13.0/velero-v1.13.0-linux-amd64.tar.gz
tar -xzf velero.tar.gz
sudo mv velero-v1.13.0-linux-amd64/velero /usr/local/bin/

Step 2 — Install Velero into the Cluster

velero install \
  --provider aws \
  --plugins velero/velero-plugin-for-aws:v1.9.0 \
  --bucket velero-backups \
  --backup-location-config region=minio,s3ForcePathStyle=true,s3Url=http://your-minio-endpoint:9000 \
  --secret-file ./credentials-velero

Even for S3-compatible (non-AWS) storage like self-hosted MinIO, Velero's AWS-compatible plugin works via the S3-compatible API, a common pattern for self-hosted backup destinations.

Step 3 — Create Your First Backup

velero backup create initial-backup

Step 4 — Check Backup Status

velero backup describe initial-backup
velero backup logs initial-backup

Step 5 — Schedule Recurring Backups

velero schedule create daily-backup --schedule="0 2 * * *"

See How to Set Up Automated Backup Verification and Alerting for the general principle applied here — manual one-off backups aren't sufficient for genuine disaster recovery preparedness; automate on a schedule.

Backing Up Specific Namespaces Only

velero backup create app-backup --include-namespaces production

Useful if you want separate backup policies for different namespaces, or don't need every namespace (some may hold genuinely ephemeral/reproducible resources not worth backing up).

Restoring from a Backup

velero restore create --from-backup initial-backup

Restores the backed-up resources and persistent volume data — test this process (see How to Test a Full Disaster Recovery Scenario (Fire Drill)) in a non-production context before relying on it during a genuine emergency.

Backing Up Persistent Volume Data

Velero can integrate with your storage provider's native snapshot capability, or use file-level backup (restic/kopia integration) depending on your setup — verify your specific persistent volume backend is properly supported for the data (not just resource definitions) to be genuinely captured.

Setting Retention for Backups

velero backup create weekly-backup --schedule="0 3 * * 0" --ttl 720h0m0s

--ttl controls how long a backup is retained before automatic deletion — balance genuine recovery needs against storage cost, similar to general backup retention considerations covered elsewhere in this Knowledge Base.

Common Errors

Backup completes but persistent volume data isn't actually included — verify your storage provider's CSI driver supports Velero's volume snapshot mechanism, or that restic/kopia file-level backup is properly configured if your storage backend doesn't support native snapshots.

Continue Reading

Browse more articles in Kubernetes & Container Orchestration.

知識庫

分類

分類